Keywords

1 Legal Reasoning as Contextual

Legal reasoning is about the creation, interpretation, application, and extinction of legal norms (rules, standards, or principles). Legislators and lawmakers argue about the creation and extinction of norms, or, more technically, about the enactment and abrogation of norms by the competent legal authorities. Judges and other officials mainly argue about the application of norms, on the basis of the evidence about the relevant facts and the interpretation of the relevant legal texts.

When the lawmaking process is legislative, participants give and ask for reasons in the parliamentary debates concerning a bill, and, more generally, in the political or scholarly debates about it. In these contexts, legal reasons frequently blend with political, economic, and moral reasons in favor of or against a bill.

When the process is judicial, reasons are given and asked for by the parties litigating before a judge or court. If this process has a lawmaking component, legal reasons will concern the merits or demerits of the new law; if it is basically on the interpretation and application of preexisting law, legal reasons will be about the correctness of certain ways of interpreting and applying it.

In some jurisdictions judges have lawmaking powers; in others they must restrain themselves and apply preexisting law, once the relevant texts have been considered and interpreted. And in most jurisdictions judges are supposed to justify their decisions. If their task is to correctly interpret and apply the law, they are supposed to provide the reasons that justify their decisions in those respects. Such reasons are usually given in written form, in official documents where judges indicate how they decided and why. Written judicial opinions articulate the reasons in favor of or against a certain decision.

Those reasons can be reconstructed as premises of reasoning schemes or patterns. In judicial contexts litigating parties provide them in the first place; then judges either accept or reject them insofar as their task is to assess the correctness of the interpretive and applicative claims of the parties; judges can also advance additional reasons when their decision-making powers go beyond the claims of the parties. Being advanced in public, in oral or written form, the reasoning of both parties and judges is a typical form of argument, properly speaking, that is the public presentation of reasons in favor of or against a litigated claim, according to logical schemes or reasoning patterns that are acceptable in the relevant context.

In the usage we adopt here, argument is wider than reasoning, and reasoning is a basic notion in that there is no argument without reasoning, while there is reasoning without argument. Some distinguishing features of an argument are that it is performed in public, that it is about a disputed point or claim, and that it is part of a dialectical exchange, where critical questions are posed and the claim is unsettled. As one scholar has put it (Walton 2018, p. 68), in an argument “the conclusion is always the claim made by one party that is doubted or is open to doubt by the other party. … Indeed, that is the whole point of using an argument. If there is no doubt about a proposition, and everybody accepts it as true, there is no reason for arguing either for or against it.”Footnote 1

Since legal reasoning in legislative contexts is usually characterized by the presentation of reasons that are not only legal (being also political, economic, or moral),Footnote 2 the most typical forms of legal reasoning occur in judicial contexts, where claims and arguments usually have a more technical outlook. This is not to deny that extralegal considerations can play a significant role in judicial decision-making; on the contrary, this is often the case. However, from the point of view of legal justification, judicial decisions must be supported by legal reasons. Economic arguments, for instance, need to be legally relevant in order to be acceptable: they must articulate reasons that are legally relevant since a legal norm takes them into explicit consideration or because they resonate with some principle or value promoted by the legal system. Welfarist arguments, in particular, can be relevant to legislative or judicial decision-making only insofar as the pursuit of welfare is considered a legal reason within the legal system. To put it differently, one thing is the law as it is (like it or not); another is the law as it ought to be, or as we wish it to be. (Economists can think of the parallel distinction between positive and normative economic analysis).

For the reasons mentioned this chapter focuses on legal reasoning as performed in judicial contexts. In a typical dispute two parties confront each other before a third one. The litigants advance claims and counterclaims, and support them with arguments and counterarguments. If the plaintiff in a civil lawsuit states that she/he was wrongfully harmed by the defendant, then she/he has to provide reasons and arguments to that effect. She/he must prove that she/he was harmed by the defendant and must explain why the harm was wrongful. Then the defendant can defend herself/himself by claiming, for instance, that there was no harm at all, or that the evidence presented to prove it is insufficient, or that the harm was caused by someone else, or that she/he had an excuse, or that the harm was not wrongful at all. If the plaintiff contends that she/he had an economic loss because of the defendant’s activity, the defendant can try to show that such loss didn’t occur, or that it is not proven, or that it was due to someone else’s activity, or that she/he caused it in order to prevent a greater harm, or that the loss was not wrongful since it was the result of a fair economic competition.

Once the arguments and counterarguments are on the table, it is the third party’s task to adjudicate the dispute. It can be an individual judge, or a court, or a jury. The decision-maker is supposed to evaluate the arguments and counterarguments presented and make a legal decision. If it is found that the plaintiff was wrongfully harmed, a decision must be made in her/his favor entitling her/him to some remedy like compensation. If instead the decision-maker finds for the defendant, no remedy will be given. A jury is not supposed to articulate the reasons why the case is decided in a given way. But judges and courts are supposed to do so in most contemporary legal systems. Judicial opinions, or reasoned decisions typically given in written form, articulate the reasoning whose ultimate conclusion is the outcome of the case.

Schematically, argumentative practice in judicial contexts encompasses evidentiary arguments that reconstruct the relevant facts of a case; interpretive arguments that extract legal rules, standards, or principles from authoritative texts; and integrative arguments that fill in the gaps in the law. At trial, the litigators support their factual and normative claims with arguments of these sorts. Evidentiary arguments are presented to support a certain version of the facts; interpretive arguments support the normative claims that a party advances given the alleged facts and the relevant authoritative texts or sources; and integrative arguments fill in the normative gaps that a legal system may present. So, legal reasoning is about these various things, namely the facts, the interpretation of texts, the filling of gaps, and the application of norms to the relevant facts.

In the following we explore some varieties of legal reasoning (Sect. 2.2) and, starting from the traditional model of the judicial syllogism, we present what we call the “double justification model” of judicial decision-making (Sect. 2.3). This means that we will address first some contents of legal reasoning and then its structure according to the syllogistic model and the double justification model.

2 The Varieties of Legal Reasoning

We won’t discuss here whether legal reasoning can be formalized or rather remains an informal practice. Artificial intelligence scholars and deontic logicians provide tools and models for the purpose of formalization. Other scholars stress the dialectical and rhetorical dimensions of legal reasoning.Footnote 3 For sure, as mentioned earlier, legal reasoning is publicly performed as an argumentative effort to persuade some audience or justify a decision. Parties primarily use it to persuade judges. Judges primarily use it to justify their decisions. Let us expand on what this is about.

In the first place, in judicial contexts, legal reasoning is about the facts of a case (see, for example, Twining 1990; Anderson et al. 2005; Haack 2014). Evidence is collected and presented to the fact-finders. But evidence by itself is not sufficient, since parties have to construct arguments out of it. It is not sufficient, for instance, that a piece of evidence like a document or a material object be shown to the fact-finders, or that a witness be brought to the witness stand. One has to articulate what the piece of evidence is supposed to prove and how it proves it, or why the witness’ testimony is credible, or why it resists the critical questions that are posed about it, and so on. The arguments used in the process of fact-finding and evidence-based inference are numerous and have varying degrees of persuasive force and justificatory power. Some traditional ones like the argument from lay testimony and the argument from documentary evidence are currently losing some of their importance whereas the arguments from scientific evidence and expert opinion are becoming more and more prominent. It is possible to provide a deductive model of reasoning on facts (Comanducci 2000), but evidentiary arguments rather exemplify nondeductive models of reasoning such as induction (Ferrer 2007), abduction (Tuzet 2003), and inference to the best explanation (Pardo and Allen 2008). Why is that? Because evidentiary arguments are constructed and received under factual uncertainty (Redmayne 2006).

As an additional aspect, reasoning is performed when evidence is assessed to determine its probative value. According to “atomistic” models of evidence assessment, fact-finders have to consider each piece of evidence in its own right, to determine its admissibility and, if admitted, its probative value. According to “holistic” models, fact-finders need to consider the whole amount of evidence at their disposal, for it is often the case that single pieces of evidence cannot prove a claim that can be proven when taken together. Next, once the evidence is assessed, fact-finders need to consider whether it meets the relevant standard of proof. In Anglo-American legal systems, the traditional standard in criminal cases is proof beyond a reasonable doubt; in civil cases, it is the preponderance of evidence, or balance of probabilities. The criminal standard requires an amount of evidence that only leaves room for unreasonable doubts about the defendant’s guilt. If the evidence presented and assessed makes it reasonable to believe that the defendant is guilty and makes it unreasonable to doubt it, then the prosecution is entitled to a verdict in its favor and the fact-finders are committed to decide against the defendant. Instead, in civil cases the fact-finders are supposed to decide for the party whose claim is better supported by the evidence which was presented and assessed. If the plaintiff claims to have been harmed by the defendant, fact-finders are supposed to decide for the plaintiff if there is a preponderance of the evidence in favor of the plaintiff’s claim; otherwise they must decide for the defendant. In the version of the standard that explicitly uses probabilities, the party whose claim appears to be more probable given the evidence is entitled to a decision in its favor. Over the last years the literature on this topic has significantly grown. Suffice it to say that, on the one hand, qualitative formulas like “beyond a reasonable doubt” need interpretation and that, on the other hand, quantitative accounts in probability terms run the risk of artificial precision. As to the first problem, what do we mean by “reasonable” doubt? How can we tell reasonable from unreasonable doubts? With regard to the second problem, for example, how can we translate the testimony of a witness into probabilities? Using subjective probabilities is a poor solution in this domain, since parties and fact-finders are supposed to provide reasons for the assessment of evidence; they are not supposed to disclose their mere preferences or subjective probabilities.

In the second place, legal reasoning is about the norms that govern a case. These norms can be legal rules, standards, or principles. The basic problem is that decision-makers do not find norms as such. What they usually have is a bunch of authoritative texts, materials, and precedents, and the norms provided by these sources can be unclear or disputed. According to some authors, in most cases the directives of action provided by legal sources are clear, and no interpretation is needed (Marmor 2005, making a distinction between interpretation and understanding). So, at least in the cases just mentioned, judges are not required to perform a specific kind of reasoning to identify the rule of the case. They simply understand what the law says. On the contrary, others claim that legal sources are always in need of interpretation, given the complexity of legal systems, the indeterminacy of legal language, and the fact that the content of legal sources is disputable even when it is clear (see Guastini 2004; as for the disputability of a legal answer when the law is clear, see Endicott 1996). Even in the easiest cases—so the argument goes—legal practitioners make interpretative choices with regard to the sources to be considered, their content, and the circumstance that a given interpretation is suitable to the case. In this view, therefore, norms are not the input but the output of the interpretive process. The input is constituted by the provisions, materials, and precedents the process starts with.

As it may be, legal practitioners are supposed to provide arguments that justify the choice of the norm to be applied to the case. Parties provide these reasons in the first place; then judges are supposed to evaluate them, to make decisions on the litigated points, and to justify their decisions.Footnote 4 What we call first-order arguments about the interpretation and application of law are used to support these claims, both by parties and judges. The contemporary theories of legal interpretation and argumentation (e.g. Alexy 1978; Tarello 1980; MacCormick and Summers 1991; Guastini 2004; Walton et al. 2018) distinguish several such arguments, notably the followingFootnote 5:

  1. 1.

    Literal arguments, or arguments from wording

  2. 2.

    A contrario arguments, or arguments from the silence of legislature

  3. 3.

    Psychological arguments, or arguments from legislative intention

  4. 4.

    Teleological arguments, or arguments from purpose

  5. 5.

    A simili arguments, or arguments from analogy

  6. 6.

    Arguments from precedent

  7. 7.

    Systemic arguments, or arguments from systemic coherence

  8. 8.

    Arguments from principle

  9. 9.

    Arguments from equity

All these arguments extract normative content from authoritative sources, namely from legal texts that are in need of interpretation and application to actual cases. And some of the listed arguments, especially the arguments from analogy, have an integrative function, since they fill in the gaps in the law. Typically, integrative arguments point at some relevant similarities or dissimilarities between cases, under the general principle that similar cases should be treated alike and different cases should be treated differently. For instance, analogy performs this integrative role by claiming that the unregulated case is relevantly similar to a regulated one and therefore should be treated alike.

It is not always easy to distinguish one argument from another, and to appreciate its role. A significant example of this is the controversial relationship between psychological and teleological arguments (see Sartor 2002; Westerman 2010). Some legal scholars claim, in particular, that in EU law the former are less important than the latter, since EU directives are formulated in terms of goals and objectives to achieve. Some add that teleological arguments are “objective” while psychological ones are merely “subjective”. For the very same reason other scholars claim that it is not important to argue about goals, when these are already stated by legislative authorities. Pauline Westerman has claimed, in particular, that “goal-regulation can be understood as a complete reversal of the traditional state of affairs, in which rules fix and prescribe a certain course of action to be followed in order to reach a certain goal. … In goal-regulation that relationship is reversed. The goals are fixed and the means are left undetermined” (Westerman 2010, p. 216). In her opinion, “most of the teleological interpretation necessarily turns into historical interpretation, focusing on the aims and purposes of the various legislators involved. This limitation affects the kind of arguments that are put forward as justification for decisions. Only explicit aims have justificatory power” (Westerman 2010, p. 222). If so, arguments from purpose convert into arguments from intention (or “historical” arguments). But one may still claim that the very point of such legislative efforts is to focus on goals rather than on intentions.

Arguments from economic consequences are of special interest for law and economics scholars. The basic idea is to justify (or criticize) a decision on the basis of its economic consequences, actual or expected. However, in legal practice such arguments are usually presented under different headings, as arguments from (economic) purpose or arguments from (economic) intention (see Cserne 2020). This is not surprising when one realizes that, to have justificatory power in law, such consequences must be legally relevant (Esposito and Tuzet 2020)—namely, relevant to a legislative purpose or intent, or relevant to the implementation of a policy or the promotion of a principle.Footnote 6 Of course, there is more room for economic considerations in lawmaking activities such as legislation.

One has also to consider the argumentative structures that parties and judges use to justify their claims. Almost always more than one argument is used. Then arguments are arranged in convergence- or chain-structures. In the former some independent arguments lead to the same conclusion. In the latter the conclusion of one argument is a premise of another. It is an empirical matter whether parties and judges reason more often according to convergence- or chain-structures. For sure, one advantage of a convergence-structure is that the conclusion may still hold in case one of the convergent arguments is rejected.

Unfortunately, it is often the case that one argument pulls in one direction and another in a different one. For instance, arguments from wording and from purpose frequently conflict in hard cases. Then decision-makers need second-order arguments which provide preference rules employed to prefer an argument over another. Such preference rules are based on normative conceptions of interpretation as value-oriented (Wróblewski 1992, pp. 61ff). The need for those rules follows from the fact that very frequently the conflicting standpoints are supported by different arguments. In principle, for any legal argument there is a possible counterargument (Llewellyn 1950). As a typical controversy, one party advocates a literal interpretation of a normative text and the other party contends that the text must be interpreted in a purposive way. If such arguments lead to opposite conclusions, the decision-maker needs a reason to prefer one over another.

Interestingly, Richard Posner has claimed that economic considerations can play a significant role when judges have discretion : courts can be legitimately guided by economic considerations (wealth maximization in Posner’s own view) “where the Constitution or legislation does not deprive them of initiative or discretion in the matter” (1985, p. 103). This counts as using economic arguments as first-order ones when preexisting law does not rule the matter.Footnote 7 But similarly, judges can use economic arguments as second-order ones when first-order arguments conflict and the economic considerations can tip the scales in favor of one of the first-order arguments.

It is sensible to claim that in criminal law textual or literal arguments should prevail over others because they put more constraints on judicial interpretation and decision-making and therefore better protect the rights of criminal defendants. If a legal system has an interpretive directive that dictates the preference for one argument over another, such a directive can be used as a second-order argument, whereas the arguments in the ranking are first-order ones. Second-order arguments apply the systemic preference criteria about first-order ones. More precisely, they concern the precedence of some argument when they require it to be used before others; and they concern the prevalence of it when they require that, in case of conflict between outcomes generated by different arguments, one argument be given more weight or strength. The precedence relation is usually accompanied by the idea that the subsequent arguments need not come into play if the precedent ones are sufficient to settle the issue.

On the one hand, some legal systems provide explicit lists of first-order arguments (e.g. art. 12 of the preliminary provisions of the 1942 Italian Civil Code, art. 3 of the 1889 Spanish Civil Code, § 7 of the Austrian “General Civil Code” of 1811). On the other hand, it is very hard to find in positive law explicit indications of second-order arguments. One may claim that, in criminal law, a strict literal interpretation should in general prevail over other arguments and considerations. But one can also find several cases where the latter are found to prevail (see, for example, Canale and Tuzet 2017). There are tendencies in legal practice that amount to implicit second-order arguments, and as such those tendencies are susceptible to many exceptions depending on the specific context and stakes.

To give an example, if the law considers as an aggravating circumstance of an offence the “use of a firearm” during and in relation to a drug-trafficking crime, does this encompass the exchange of an automatic weapon for cocaine? In Smith v. U.S. (U.S. Supreme Court, 1993) the starting argument against the defendant was that “use of a firearm” literally encompasses any use of a firearm that facilitates the commission of a drug offence, including the use of it as an item of barter. Against this argument the defense pointed out that, in a contextual understanding of language, when we refer to the use of an artifact we refer to the standard or intended use of it (the use the artifact was created for). Therefore, so the argument went, the relevant “use of a firearm” would be the use of it as a weapon and in the case in hand the defendant didn’t use it as such, for he tried to employ it as a means of payment. Against the defendant it was also pointed out that the legislative purpose was to minimize the risk that the presence of drugs and firearms imposes on individuals and society. Drugs and guns are a dangerous combination and, the argument went, any use of a firearm during and in relation to a drug-trafficking crime should be sanctioned in order to minimize that risk. Now, the argumentation theorist can apply to this case the categories mentioned earlier and claim that the convergent combination of the first literal argument and the argument from purpose outweighed, in the final decision, the second literal argument, namely the argument from contextual meaning advanced in favor of the defendant. The defendant was sentenced to a significant prison term.

3 A Model and Its Enhancement: The Judicial Syllogism and The Double Justification Model

Analytic theorists typically divide legal arguments into two broad categories: arguments about facts and arguments about norms. Arguments about facts, as already pointed out, aim at justifying (or contesting) the reconstruction of the relevant facts. Parties typically produce evidence to support their factual claims, or to contest rival claims. And given that evidence per se doesn’t yield verdicts, the evidence presented is in need of being “inferentialized”, that is translated into evidentiary inferences and arguments aimed at persuading the fact-finders. Arguments about norms aim at justifying (or contesting) the identification and application of legal norms. Parties typically discuss about a legal provision which one of them at least considers to be relevant and applicable to the case in hand. The provision is usually in need of interpretation, or of being contextualized to the system or subsystem it belongs to. The latter operation requires a reconstruction of the system, of its parts, principles, statutory norms, relevant judicial precedents, and so on.

Once the facts are found and the relevant norms are identified, the decision-makers need only apply the latter to the former. For some analytic approaches this kind of application consists, logically speaking, in a deductive inference. According to the traditional model of the judicial syllogism, decision-makers are to deduce the outcome of the case from the facts and the applicable norms. The model was advanced by Cesare Beccaria in his 1764 masterpiece Dei delitti e delle pene (Chap. IV). It was meant as a normative model to constrain judicial decision-making in criminal law. It was not presented as a model descriptive of judicial practice. In fact, Beccaria was quite critical of the criminal justice system of his time. His main critique concentrated on the arbitrariness of criminal decision-making in his days. As a remedy to it, he recommended that judges decide according to a syllogism model, with a general legal norm provided by legislation as major premise, the relevant fact as minor premise, and the outcome as a logical deductive conclusion.

To make a very simple illustration (of course actual cases are far more complex), if legislation establishes that whoever does A shall be punished with S (major premise), and if Basil did A (minor premise), then Basil shall be punished with S (conclusion). The argument has a deductive logical structure: if the premises are true, the conclusion cannot be false. Or, the conclusion is necessarily correct given the correctness of the premises. The model preserves legal certainty (or the rule of law), as well as the principle of equal treatment under the law: if Basil did A and whoever does A shall be punished with S, not punishing him with S would be to treat him differently. Deductive application of law permits to treat like cases alike.

The model can be extended from criminal to civil matters. If legislation establishes that whoever wrongfully harms another person shall compensate this person (major premise), and Basil wrongfully harmed Theodor (minor premise), then Basil shall compensate Theodor (conclusion). Again, the argument has a deductive logical structure in that the conclusion cannot be false if the premises are true. And again, it preserves legal certainty and equality. Both logical and legal principles support the model.

But judges need reasons to assume the premises of the syllogism. They do not find the premises as one finds mushrooms under the trees. The premises must be determined out of the relevant legal sources and evidence. Parties present arguments whose conclusions are possible premises of the judicial syllogism. Then adjudicators evaluate their arguments and, if needed, supplement them. So, as many critics have pointed out, the syllogism model is too simple in this respect. One needs arguments for the premises. In fact, the vast majority of legal disputes concern such arguments.

That calls for an enhancement of the model. A well-established view in the contemporary literature has it that the justification of judicial decisions is double: internal and external. We call “internal” the justification of the conclusion provided by the deductive structure of the syllogism, and “external” the justification of its premises.Footnote 8 On the whole, we can call this the double justification model.

In turn, external justification has two aspects. Interpretive and integrative arguments provide the external justification of the major premise of the syllogism, and evidentiary arguments provide the external justification of its minor premise. Both aspects, normative and factual external justification, may be determined by first- and second-order arguments. As simple illustrations, imagine a context where literal meaning prevails over purposive meaning and a context where scientific testimony prevails over a lay one. In such contexts some second-order arguments dispose of the conflicts between first-order ones. Then, once the premises are established along such lines, the syllogism provides the internal justification of the outcome. If it has been normatively established that whoever wrongfully harms another person shall compensate this person, and it has been proven that Basil wrongfully harmed Theodor, then Basil shall compensate Theodor. The conclusion has a double justification if both premises and conclusion of the reasoning are justified.

In its typical structure, normative external justification goes from provisions to norms through interpretive (or integrative) arguments (see § 2 earlier). In its typical structure, factual external justification goes from evidence to factual reconstruction through “bridge rules”. Bridge rules are empirical generalizations, scientific laws, and legal rules concerning the probative value of the evidence. Bridge rules connect the evidence at disposal with the relevant facts as they can be reconstructed (again, see § 2 earlier). Jerzy Wróblewski (1971, p. 415) distinguishes in this respect “rules of empirical evidence” and “rules of legal evidence”, to account for the various ways (our “bridge rules”) in which evidence leads to the facts to be proven; a rule of the first kind can be a rule of common sense, a rule of the second kind a legal presumption. In a partially different account, bridge rules connect secondary to primary facts. Secondary facts are the probatory facts evidence amounts to (e.g. the fact that witness W said that p); primary facts are the facts to be proven, namely the legally relevant facts (whether it was the case that p); and bridge rules justify the reasoning from the ones to the others (being W a reliable witness, it can be inferred that p). Fact-finders assess the evidence along such lines and decide according to the standard of proof.

From a logical point of view, one can conceive of normative external justification as deductive if one considers interpretive arguments as directives that, like normative major premises, shall be applied to the interpretive problem at hand (then one has to solve, with second-order arguments, the possible conflicts between such directives). Or, one can look at normative external justification as the epistemic effort to find the best interpretation for the case in hand. This is possible if one considers interpretive arguments as heuristic devices that help interpreters find the correct or best interpretation of the relevant materials for the case in hand. Logically speaking, in this sense, normative external justification becomes abductive, or a form of inference to the best explanation, being an educated guess at what is most correct in legal terms. Factual external justification is more straightforwardly epistemic, even if some bridge rules have a legal nature and the standards of proof respond to principles and values. From a logical point of view, factual external justification is mainly abductive; or, more generally speaking, it is the effort to find the best explanation of the evidence and to check whether it satisfies the relevant standard of proof.